Librem5 issueshttps://source.puri.sm/groups/Librem5/-/issues2019-09-26T15:56:13Zhttps://source.puri.sm/Librem5/design/-/issues/22Adaptive Shell2019-09-26T15:56:13ZTobias BernardAdaptive ShellWe need a plan for how Phosh will scale to larger screens.
Some preliminary goals:
- align with GNOME Shell philosophically and structurally (distraction free, single path, rich search, etc.)
- seamlessly scale from phone to desktop
- c...We need a plan for how Phosh will scale to larger screens.
Some preliminary goals:
- align with GNOME Shell philosophically and structurally (distraction free, single path, rich search, etc.)
- seamlessly scale from phone to desktop
- clean spatial model across session, lock, and login (and ideally boot/shutdown too)
- avoid "workflow potholes" (infinite windows per workspace etc.)
- great keyboard-only workflow
- great touch-only workflow
Prior art:
- iPad OS tiling and workspaces
- tiling window managers (i3 and the like)
- webOS cards
- macOS fullscreen workspaces
- Windows suggested apps when tiling
- iOS/Android multitasking gestures
Ideas:
- tiling-first (no floating windows, with some well-contained exceptions)
- horizontal workspaces with app grid below them, similar to mobile
- alt-tab integrated with shell overview (one-handed quick keyboard switching)
Maybe somewhat related: https://source.puri.sm/Librem5/design/issues/21https://source.puri.sm/Librem5/image-builder/-/issues/44maybe switch to f2fs2021-03-09T13:19:25ZGuido Gunthermaybe switch to f2fsswitching to f2fs from ext4 might make sense: https://en.wikipedia.org/wiki/F2FS
- there's a f2fs f2fscrypt
- resize.f2fs exists so we should be able to make or resize on first boot workswitching to f2fs from ext4 might make sense: https://en.wikipedia.org/wiki/F2FS
- there's a f2fs f2fscrypt
- resize.f2fs exists so we should be able to make or resize on first boot workhttps://source.puri.sm/Librem5/gnome-control-center/-/issues/68hide parts of 'notifications'?2019-09-25T15:53:40ZGuido Guntherhide parts of 'notifications'?we currently don't provide notifications on the lock screen https://source.puri.sm/Librem5/phosh/issues/166 nor per appliction config https://source.puri.sm/Librem5/phosh/issues/168
we could temporarily hide these but also postpone this...we currently don't provide notifications on the lock screen https://source.puri.sm/Librem5/phosh/issues/166 nor per appliction config https://source.puri.sm/Librem5/phosh/issues/168
we could temporarily hide these but also postpone this a little and see what we get slipped soonish.https://source.puri.sm/Librem5/gnome-control-center/-/issues/67time zone selection does not touch scroll2020-02-26T09:53:25ZGuido Gunthertime zone selection does not touch scrolli can move he scroll bar but it would be cool if i could shift around the map via touch as well.i can move he scroll bar but it would be cool if i could shift around the map via touch as well.https://source.puri.sm/Librem5/gnome-control-center/-/issues/64devices -> mouse and touch pad toggles too wide window2021-07-26T14:58:15ZGuido Guntherdevices -> mouse and touch pad toggles too wide windowto reproduce
- start gnome-control-center
- hit devices
- hit mouse and touchpad (window does not fit)
- hit back arrow in top left corner
- hit back arrow in top left corner again
see the window being too wide displaying parts of the ...to reproduce
- start gnome-control-center
- hit devices
- hit mouse and touchpad (window does not fit)
- hit back arrow in top left corner
- hit back arrow in top left corner again
see the window being too wide displaying parts of the cellular dialog, it's basically already visiable after hitting back once but pressing two times makes it very obvioushttps://source.puri.sm/Librem5/gnome-control-center/-/issues/62Dialog for testing audio doesn't fit2019-09-26T08:07:21ZJulian SparberDialog for testing audio doesn't fit![audio](/uploads/1412d78a7644089e29df2064f7be7a7b/audio.jpeg)![audio](/uploads/1412d78a7644089e29df2064f7be7a7b/audio.jpeg)https://source.puri.sm/Librem5/gnome-control-center/-/issues/61slow startup2021-06-09T07:18:52ZGuido Guntherslow startupg-c-c's startup is pretty slow, @mohammed.sadiq says there are sync dbus calls that could cause that (which makes sense since e.g. starting twice in a row doesn't make it much faster). needs investigation to make the whole ui of the phon...g-c-c's startup is pretty slow, @mohammed.sadiq says there are sync dbus calls that could cause that (which makes sense since e.g. starting twice in a row doesn't make it much faster). needs investigation to make the whole ui of the phone snappy.https://source.puri.sm/Librem5/image-builder/-/issues/42santize version to legacy/current/next mapping2019-09-24T14:10:56ZGuido Gunthersantize version to legacy/current/next mappingcurrently we have
```
case "${build_type}" in
legacy)
kernel_url="${kernel_base}/kernel_devkit_linux-next_imx8-legacy-devkit/${archive_path}"
uboot_url="${uboot_base}/uboot_devkit_mainline_atf_build/${archive_pa...currently we have
```
case "${build_type}" in
legacy)
kernel_url="${kernel_base}/kernel_devkit_linux-next_imx8-legacy-devkit/${archive_path}"
uboot_url="${uboot_base}/uboot_devkit_mainline_atf_build/${archive_path}"
;;
next)
kernel_url="${kernel_base}/kernel_devkit_linux-next_imx8-linux-next-devkit/${archive_path}"
uboot_url="${uboot_base}/uboot_devkit_mainline_atf_build/${archive_path}"
;;
current)
kernel_url="${kernel_base}/kernel_devkit_linux-next_imx8-current-devkit/${archive_path}"
uboot_url="${uboot_base}/uboot_devkit_mainline_atf_build/${archive_path}"
;;
*)
echo "unknown kernel type for devkit build ${build_type}"
exit 1
;;
esac
```
i would opt for having version numbers in the actual jobs and do the mapping to legacy/next/current in `conf/board`. this allows us to have different, easy to identify build jobs and only one place - here - that maps to `legacy/next/current` making it simple to change without changing under them, so e.g.
```
kernel_url="${kernel_base}/kernel_devkit_linux-next_imx8-5.3-devkit/${archive_path}"
```
instead of
```
kernel_url="${kernel_base}/kernel_devkit_linux-next_imx8-current-devkit/${archive_path}"
```
also if we build for phone and devkit we should drop `devkit` from job names.https://source.puri.sm/Librem5/debs/gtk/-/issues/9touch events with finger down2022-10-03T15:45:48ZGuido Gunthertouch events with finger downI often end up holding the devkit/phone with one hand slightly touching the display in an area where there's no widget.
When i then press e.g. a button that won't do anything since the first finger is already interpreted as a touch eve...I often end up holding the devkit/phone with one hand slightly touching the display in an area where there's no widget.
When i then press e.g. a button that won't do anything since the first finger is already interpreted as a touch event. to reproduce:
- boot the devkit
- slide up the lock screen
- press with one finger on the bottom of the screen below the keypad
- press one of the numbers with another finger
result: nothing happens
expected result: the touch of the first finger is ignored and the second finger touch of the number is detected
somewhat related https://source.puri.sm/Librem5/squeekboard/issues/33https://source.puri.sm/Librem5/linux/-/issues/90Redpine module stays in reset when nSDIO_RST high2019-11-12T19:37:20ZAngus Ainslieangus.ainslie@puri.smRedpine module stays in reset when nSDIO_RST highPin 23 of the M.2 connector is SDIO_RST# according to the M.2 spec. The spec specifies SDIO_RST# as a 1.8V input. However, from testing, the Redpine module only comes out of reset if this pin is above 2.3V.
Here is an excerpt from the s...Pin 23 of the M.2 connector is SDIO_RST# according to the M.2 spec. The spec specifies SDIO_RST# as a 1.8V input. However, from testing, the Redpine module only comes out of reset if this pin is above 2.3V.
Here is an excerpt from the specification:
![image](/uploads/5ebe81f7045c3d0faf46ff1f1ddaf15c/image.png)https://source.puri.sm/Librem5/gnome-initial-setup/-/issues/8needs a mobile broadband page2019-11-19T13:56:35ZGuido Guntherneeds a mobile broadband pagewe should have a page that allows to select e.g.the apn for mobile data connectionwe should have a page that allows to select e.g.the apn for mobile data connectionhttps://source.puri.sm/Librem5/gnome-initial-setup/-/issues/6during initial boot g-i-s should start over an unlocked shell2020-01-21T00:00:04ZGuido Guntherduring initial boot g-i-s should start over an unlocked shellthis likely needs some interaction with the shell we could
- starting phosh initally with `-U`,
- the user should then be able to set up a pin via gis
- and the phone will then boot up locked in the future
does this make sense?this likely needs some interaction with the shell we could
- starting phosh initally with `-U`,
- the user should then be able to set up a pin via gis
- and the phone will then boot up locked in the future
does this make sense?https://source.puri.sm/Librem5/Apps_Issues/-/issues/159Need ways to detect adaptive apps from desktop file2021-08-03T15:37:10ZGuido GuntherNeed ways to detect adaptive apps from desktop filewhen thinking about adaptive we want to have
1. apps installed that work in desktop-mode, e.g. libre office
2. apps installed that work in both desktop and phone mode: epiphany, chatty, calls, ...
3. maybe have apps that are limited to ...when thinking about adaptive we want to have
1. apps installed that work in desktop-mode, e.g. libre office
2. apps installed that work in both desktop and phone mode: epiphany, chatty, calls, ...
3. maybe have apps that are limited to phone mode
identifying these application types allows us to show clean menus in desktop and phone mode.
Since there's no point in showing apps from 1. with no external screen attached we need to somehow tag adaptive apps to avoid clutter like https://source.puri.sm/Librem5/librem5-base/merge_requests/65 in the future.
One way would be to tag convergent apps with `X-Adaptive: true` in the .desktop file and only these show up in the menu in phone mode. Possible problems with that:
- webapps from epiphany won't have that tag (but that could be fixed)
- kirigami and apps based on other toolkits would need to use that as well
we might end up having to whitelist some apps until upstream catches up patch downstream. in the long run we want to extend the freedesktop.org spec i assume.
/cc @adrien.plazas @tobias.bernard @francois.techene if this makes sense in their plans of convergence?
/cc @dorota.czaplejewicz @david.boddie since this might have been a topic in dicussions with plamo?https://source.puri.sm/Librem5/gnome-control-center/-/issues/59Power: weird vertical padding on list rows (VM image only)2019-09-20T13:50:04ZTobias BernardPower: weird vertical padding on list rows (VM image only)This only happens on the VM image in Boxes, but I thought it would be good to track it anyway:
![image](/uploads/d89cac5e8c327e00c3a3f3760ae54bce/image.png)This only happens on the VM image in Boxes, but I thought it would be good to track it anyway:
![image](/uploads/d89cac5e8c327e00c3a3f3760ae54bce/image.png)https://source.puri.sm/Librem5/Apps_Issues/-/issues/158usb gadget / host mode switch2020-06-13T01:45:29ZGuido Guntherusb gadget / host mode switchWe want a simple switch between usb host and gadget mode - most of the users want host host mode i assume so they but for flashing/debugging gadget mode is very useful. being able to switch that at run time would be great.
/cc @tobias....We want a simple switch between usb host and gadget mode - most of the users want host host mode i assume so they but for flashing/debugging gadget mode is very useful. being able to switch that at run time would be great.
/cc @tobias.bernard since we want a common place for such phone specific things i'd assumehttps://source.puri.sm/Librem5/Apps_Issues/-/issues/157packagekit updates package lists over slow (and maybe expensive) modem connec...2019-11-05T15:33:01ZGuido Guntherpackagekit updates package lists over slow (and maybe expensive) modem connectionsThe update of the Debian package lists currently eats bandwidth of already slow modem connections. this should not happen.
/cc @adrien.plazas since this was pulled in by gnome-softwareThe update of the Debian package lists currently eats bandwidth of already slow modem connections. this should not happen.
/cc @adrien.plazas since this was pulled in by gnome-softwarehttps://source.puri.sm/Librem5/gnome-control-center/-/issues/54Devices: Display panel doesn't fit with 2 monitores2020-09-14T17:36:08ZJulian SparberDevices: Display panel doesn't fit with 2 monitoresI didn't try it on the devkit but there should be the same issue. If the user connects more then one display the the row showing the display mode doesn't fit anymore.
![image](/uploads/c038290b6e96758c7cafc7ba42bb021f/image.png)I didn't try it on the devkit but there should be the same issue. If the user connects more then one display the the row showing the display mode doesn't fit anymore.
![image](/uploads/c038290b6e96758c7cafc7ba42bb021f/image.png)https://source.puri.sm/Librem5/gnome-control-center/-/issues/49Bluetooth panel (empty state) doesn't fit2021-03-30T13:42:10ZTobias BernardBluetooth panel (empty state) doesn't fitI've not been able to test this panel with working Bluetooth hardware yet, but the empty state doesn't fit:
![image](/uploads/aa99f9bcd76f7f00ae510783a3dc6d26/image.png)
Seeing as we probably won't have direct support for Bluetooth soo...I've not been able to test this panel with working Bluetooth hardware yet, but the empty state doesn't fit:
![image](/uploads/aa99f9bcd76f7f00ae510783a3dc6d26/image.png)
Seeing as we probably won't have direct support for Bluetooth soon anyway, and the rest of the panel UI will also need work to be fully adaptive, I think we should just hide it for now.https://source.puri.sm/Librem5/gnome-control-center/-/issues/45G-c-c doesn't start via ssh/serialmonitor2020-02-24T10:30:01ZJulian SparberG-c-c doesn't start via ssh/serialmonitorIt looks like that a assert fails.
Connecting via ssh to the librem5 devkit:
```
✗ ssh purism@192.168.1.5
purism@192.168.1.5's password:
Linux pureos 5.2.0-gef814e5c7 #1 SMP PREEMPT Mon Jul 29 07:35:31 PDT 2019 aarch64
The programs inc...It looks like that a assert fails.
Connecting via ssh to the librem5 devkit:
```
✗ ssh purism@192.168.1.5
purism@192.168.1.5's password:
Linux pureos 5.2.0-gef814e5c7 #1 SMP PREEMPT Mon Jul 29 07:35:31 PDT 2019 aarch64
The programs included with the Debian GNU/Linux system are free software;
the exact distribution terms for each program are described in the
individual files in /usr/share/doc/*/copyright.
Debian GNU/Linux comes with ABSOLUTELY NO WARRANTY, to the extent
permitted by applicable law.
Last login: Thu Sep 19 09:13:20 2019
purism@pureos:~$ G_MESSAGES_DEBUG=all gnome-control-center
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.953: watch_fast: "/org/gnome/desktop/interface/" (establishing: 0, active: 0)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.954: watch_fast: "/org/gnome/settings-daemon/peripherals/mouse/" (establishing: 0, active: 0)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.954: watch_fast: "/org/gnome/desktop/sound/" (establishing: 0, active: 0)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.954: watch_fast: "/org/gnome/desktop/privacy/" (establishing: 0, active: 0)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.955: watch_fast: "/org/gnome/desktop/wm/preferences/" (establishing: 0, active: 0)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.955: watch_fast: "/org/gnome/settings-daemon/plugins/xsettings/" (establishing: 0, active: 0)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.955: watch_fast: "/org/gnome/desktop/a11y/" (establishing: 0, active: 0)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.958: watch_established: "/org/gnome/desktop/interface/" (establishing: 1)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.959: watch_established: "/org/gnome/settings-daemon/peripherals/mouse/" (establishing: 1)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.959: watch_established: "/org/gnome/desktop/sound/" (establishing: 1)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.959: watch_established: "/org/gnome/desktop/privacy/" (establishing: 1)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.959: watch_established: "/org/gnome/desktop/wm/preferences/" (establishing: 1)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.959: watch_established: "/org/gnome/settings-daemon/plugins/xsettings/" (establishing: 1)
(gnome-control-center:1076): dconf-DEBUG: 09:15:12.960: watch_established: "/org/gnome/desktop/a11y/" (establishing: 1)
(gnome-control-center:1076): Gdk-DEBUG: 09:15:13.101: Compositor prefers decoration mode 'client'
(gnome-control-center:1076): Clutter-CRITICAL **: 09:15:13.345: Unable to initialize Clutter: Unable to initialize the Clutter backend: no available drivers found.
(gnome-control-center:1076): cc-object-storage-DEBUG: 09:15:13.347: Initializing object storage
(gnome-control-center:1076): Gtk-DEBUG: 09:15:13.351: Connecting to session manager
(gnome-control-center:1076): dconf-DEBUG: 09:15:13.453: watch_fast: "/org/gnome/control-center/" (establishing: 0, active: 0)
(gnome-control-center:1076): dconf-DEBUG: 09:15:13.456: watch_established: "/org/gnome/control-center/" (establishing: 1)
(gnome-control-center:1076): network-cc-panel-DEBUG: 09:15:13.487: Monitoring NetworkManager for Wi-Fi devices
(gnome-control-center:1076): cc-object-storage-DEBUG: 09:15:13.716: Adding object NMClient (CcObjectStorage::nm-client → 0xaaaae4d769c0) to the storage
**
ERROR:../shell/cc-shell-model.c:458:cc_shell_model_set_panel_visibility: assertion failed: (valid)
Aborted (core dumped)
purism@pureos:~$
```https://source.puri.sm/Librem5/gnome-control-center/-/issues/42panels sometimes start too wide2022-01-16T22:05:29ZGuido Guntherpanels sometimes start too widei can reproduce this with details->users on the devkit which always ends up a bit too wide but then when i click in a text field gets the correct width.
/cc: @julian.sparberi can reproduce this with details->users on the devkit which always ends up a bit too wide but then when i click in a text field gets the correct width.
/cc: @julian.sparber